NVIDIA · 13ernkastel · Apr 3, 2026 · Apr 3, 2026 · Apr 3, 2026 · Apr 5, 2026
diff --git a/.agents/skills/nemoclaw-configure-inference/references/inference-options.md b/.agents/skills/nemoclaw-configure-inference/references/inference-options.md
@@ -19,15 +19,15 @@ The onboard wizard presents the following provider options by default.
 The first six are always available.
 Ollama appears when it is installed or running on the host.
 
-| Option | Description | Curated models |
-|--------|-------------|----------------|
-| NVIDIA Endpoints | Routes to models hosted on [build.nvidia.com](https://build.nvidia.com). You can also enter any model ID from the catalog. Set `NVIDIA_API_KEY`. | Nemotron 3 Super 120B, Kimi K2.5, GLM-5, MiniMax M2.5, GPT-OSS 120B |
-| OpenAI | Routes to the OpenAI API. Set `OPENAI_API_KEY`. | `gpt-5.4`, `gpt-5.4-mini`, `gpt-5.4-nano`, `gpt-5.4-pro-2026-03-05` |
-| Other OpenAI-compatible endpoint | Routes to any server that implements `/v1/chat/completions`. The wizard prompts for a base URL and model name. Works with OpenRouter, LocalAI, llama.cpp, or any compatible proxy. Set `COMPATIBLE_API_KEY`. | You provide the model name. |
-| Anthropic | Routes to the Anthropic Messages API. Set `ANTHROPIC_API_KEY`. | `claude-sonnet-4-6`, `claude-haiku-4-5`, `claude-opus-4-6` |
-| Other Anthropic-compatible endpoint | Routes to any server that implements the Anthropic Messages API (`/v1/messages`). The wizard prompts for a base URL and model name. Set `COMPATIBLE_ANTHROPIC_API_KEY`. | You provide the model name. |
-| Google Gemini | Routes to Google's OpenAI-compatible endpoint. Set `GEMINI_API_KEY`. | `gemini-3.1-pro-preview`, `gemini-3.1-flash-lite-preview`, `gemini-3-flash-preview`, `gemini-2.5-pro`, `gemini-2.5-flash`, `gemini-2.5-flash-lite` |
-| Local Ollama | Routes to a local Ollama instance on `localhost:11434`. NemoClaw detects installed models, offers starter models if none are present, pulls and warms the selected model, and validates it. | Selected during onboarding. For more information, refer to Use a Local Inference Server (see the `nemoclaw-configure-inference` skill). |
+| Option | Status | Description | Curated models |
+|--------|--------|-------------|----------------|
+| NVIDIA Endpoints | Supported | Routes to models hosted on [build.nvidia.com](https://build.nvidia.com). You can also enter any model ID from the catalog. Set `NVIDIA_API_KEY`. | Nemotron 3 Super 120B, Kimi K2.5, GLM-5, MiniMax M2.5, GPT-OSS 120B |
+| OpenAI | Supported | Routes to the OpenAI API. Set `OPENAI_API_KEY`. | `gpt-5.4`, `gpt-5.4-mini`, `gpt-5.4-nano`, `gpt-5.4-pro-2026-03-05` |
+| Other OpenAI-compatible endpoint | Supported | Routes to any server that implements `/v1/chat/completions`. The wizard prompts for a base URL and model name. Works with OpenRouter, LocalAI, llama.cpp, or any compatible proxy. Set `COMPATIBLE_API_KEY`. | You provide the model name. |
+| Anthropic | Supported | Routes to the Anthropic Messages API. Set `ANTHROPIC_API_KEY`. | `claude-sonnet-4-6`, `claude-haiku-4-5`, `claude-opus-4-6` |
+| Other Anthropic-compatible endpoint | Supported | Routes to any server that implements the Anthropic Messages API (`/v1/messages`). The wizard prompts for a base URL and model name. Set `COMPATIBLE_ANTHROPIC_API_KEY`. | You provide the model name. |
+| Google Gemini | Supported | Routes to Google's OpenAI-compatible endpoint. Set `GEMINI_API_KEY`. | `gemini-3.1-pro-preview`, `gemini-3.1-flash-lite-preview`, `gemini-3-flash-preview`, `gemini-2.5-pro`, `gemini-2.5-flash`, `gemini-2.5-flash-lite` |
+| Local Ollama | Caveated | Routes to a local Ollama instance on `localhost:11434`. NemoClaw detects installed models, offers starter models if none are present, pulls and warms the selected model, and validates it. | Selected during onboarding. For more information, refer to Use a Local Inference Server (see the `nemoclaw-configure-inference` skill). |
 
 ## Experimental Options
 

diff --git a/.agents/skills/nemoclaw-overview/SKILL.md b/.agents/skills/nemoclaw-overview/SKILL.md
@@ -88,14 +88,25 @@ By combining powerful open source models with built-in safety measures, NemoClaw
 
 NemoClaw provides the following capabilities on top of the OpenShell runtime.
 
-| Feature | Description |
-|---------|-------------|
-| Guided onboarding | Validates credentials, selects providers, and creates a working sandbox in one command. |
-| Hardened blueprint | A security-first Dockerfile with capability drops, least-privilege network rules, and declarative policy. |
-| State management | Safe migration of agent state across machines with credential stripping and integrity verification. |
-| Messaging bridges | Host-side processes that connect Telegram, Discord, and Slack to the sandboxed agent. |
-| Routed inference | Provider-routed model calls through the OpenShell gateway, transparent to the agent. Supports NVIDIA Endpoints, OpenAI, Anthropic, Google Gemini, and local Ollama. |
-| Layered protection | Network, filesystem, process, and inference controls that can be hot-reloaded or locked at creation. |
+:::{list-table}
+:header-rows: 1
+
+* - Feature
+  - Description
+* - Guided onboarding
+  - Validates credentials, selects providers, and creates a working sandbox in one command.
+* - Hardened blueprint
+  - A security-first Dockerfile with capability drops, least-privilege network rules, and declarative policy.
+* - State management
+  - Safe migration of agent state across machines with credential stripping and integrity verification.
+* - Messaging bridges
+  - Host-side processes that connect Telegram, Discord, and Slack to the sandboxed agent.
+* - Routed inference
+  - Provider-routed model calls through the OpenShell gateway, transparent to the agent.
+    Supports NVIDIA Endpoints, OpenAI, Anthropic, Google Gemini, compatible OpenAI or Anthropic endpoints, and the caveated Local Ollama path.
+* - Layered protection
+  - Network, filesystem, process, and inference controls that can be hot-reloaded or locked at creation.
+:::
 
 ## Challenge
 

diff --git a/.agents/skills/nemoclaw-overview/references/overview.md b/.agents/skills/nemoclaw-overview/references/overview.md
@@ -17,14 +17,25 @@ By combining powerful open source models with built-in safety measures, NemoClaw
 
 NemoClaw provides the following capabilities on top of the OpenShell runtime.
 
-| Feature | Description |
-|---------|-------------|
-| Guided onboarding | Validates credentials, selects providers, and creates a working sandbox in one command. |
-| Hardened blueprint | A security-first Dockerfile with capability drops, least-privilege network rules, and declarative policy. |
-| State management | Safe migration of agent state across machines with credential stripping and integrity verification. |
-| Messaging bridges | Host-side processes that connect Telegram, Discord, and Slack to the sandboxed agent. |
-| Routed inference | Provider-routed model calls through the OpenShell gateway, transparent to the agent. Supports NVIDIA Endpoints, OpenAI, Anthropic, Google Gemini, and local Ollama. |
-| Layered protection | Network, filesystem, process, and inference controls that can be hot-reloaded or locked at creation. |
+:::{list-table}
+:header-rows: 1
+
+* - Feature
+  - Description
+* - Guided onboarding
+  - Validates credentials, selects providers, and creates a working sandbox in one command.
+* - Hardened blueprint
+  - A security-first Dockerfile with capability drops, least-privilege network rules, and declarative policy.
+* - State management
+  - Safe migration of agent state across machines with credential stripping and integrity verification.
+* - Messaging bridges
+  - Host-side processes that connect Telegram, Discord, and Slack to the sandboxed agent.
+* - Routed inference
+  - Provider-routed model calls through the OpenShell gateway, transparent to the agent.
+    Supports NVIDIA Endpoints, OpenAI, Anthropic, Google Gemini, compatible OpenAI or Anthropic endpoints, and the caveated Local Ollama path.
+* - Layered protection
+  - Network, filesystem, process, and inference controls that can be hot-reloaded or locked at creation.
+:::
 
 ## Challenge
 

diff --git a/.agents/skills/nemoclaw-reference/references/commands.md b/.agents/skills/nemoclaw-reference/references/commands.md
@@ -45,7 +45,8 @@ $ nemoclaw onboard
 ```
 
 The wizard prompts for a provider first, then collects the provider credential if needed.
-Supported non-experimental choices include NVIDIA Endpoints, OpenAI, Anthropic, Google Gemini, and compatible OpenAI or Anthropic endpoints.
+Supported provider choices include NVIDIA Endpoints, OpenAI, Anthropic, Google Gemini, and compatible OpenAI or Anthropic endpoints.
+Local Ollama is also available in the standard onboarding flow as a caveated provider path.
 Credentials are stored in `~/.nemoclaw/credentials.json`.
 The legacy `nemoclaw setup` command is deprecated; use `nemoclaw onboard` instead.
 

diff --git a/.agents/skills/nemoclaw-reference/references/inference-profiles.md b/.agents/skills/nemoclaw-reference/references/inference-profiles.md
@@ -16,18 +16,21 @@ At onboard time, NemoClaw configures:
 
 That means the sandbox knows which model family to use, while OpenShell owns the actual provider credential and upstream endpoint.
 
-## Supported Providers
-
-The following non-experimental provider paths are available through `nemoclaw onboard`.
-
-| Provider | Endpoint Type | Notes |
-|---|---|---|
-| NVIDIA Endpoints | OpenAI-compatible | Hosted models on `integrate.api.nvidia.com` |
-| OpenAI | Native OpenAI-compatible | Uses OpenAI model IDs |
-| Other OpenAI-compatible endpoint | Custom OpenAI-compatible | For compatible proxies and gateways |
-| Anthropic | Native Anthropic | Uses `anthropic-messages` |
-| Other Anthropic-compatible endpoint | Custom Anthropic-compatible | For Claude proxies and compatible gateways |
-| Google Gemini | OpenAI-compatible | Uses Google's OpenAI-compatible endpoint |
+## Provider Status
+
+The following provider paths are available through `nemoclaw onboard`.
+
+| Provider | Status | Endpoint Type | Notes |
+|---|---|---|---|
+| NVIDIA Endpoints | Supported | OpenAI-compatible | Hosted models on `integrate.api.nvidia.com` |
+| OpenAI | Supported | Native OpenAI-compatible | Uses OpenAI model IDs |
+| Other OpenAI-compatible endpoint | Supported | Custom OpenAI-compatible | For compatible proxies and gateways |
+| Anthropic | Supported | Native Anthropic | Uses `anthropic-messages` |
+| Other Anthropic-compatible endpoint | Supported | Custom Anthropic-compatible | For Claude proxies and compatible gateways |
+| Google Gemini | Supported | OpenAI-compatible | Uses Google's OpenAI-compatible endpoint |
+| Local Ollama | Caveated | Local Ollama API | Available in the standard onboarding flow when Ollama is installed or running on the host |
+| Local NVIDIA NIM | Experimental | Local OpenAI-compatible | Requires `NEMOCLAW_EXPERIMENTAL=1` and a NIM-capable GPU |
+| Local vLLM | Experimental | Local OpenAI-compatible | Requires `NEMOCLAW_EXPERIMENTAL=1` and a server already running on `localhost:8000` |
 
 ## Validation During Onboarding
 
@@ -46,7 +49,7 @@ If validation fails, the wizard does not continue to sandbox creation.
 
 ## Local Ollama
 
-Local Ollama is available in the standard onboarding flow when Ollama is installed or running on the host.
+Local Ollama is a caveated provider path available in the standard onboarding flow when Ollama is installed or running on the host.
 It uses the same routed `inference.local` pattern, but the upstream runtime runs locally instead of in the cloud.
 
 Ollama gets additional onboarding help:
@@ -68,7 +71,7 @@ If Ollama only binds loopback, NemoClaw can detect it on the host, but the sandb
 
 ## Experimental Local Providers
 
-The following local providers require `NEMOCLAW_EXPERIMENTAL=1`:
+The following local providers remain experimental and require `NEMOCLAW_EXPERIMENTAL=1`:
 
 - Local NVIDIA NIM (requires a NIM-capable GPU)
 - Local vLLM (must already be running on `localhost:8000`)

diff --git a/README.md b/README.md
@@ -54,13 +54,15 @@ The sandbox image is approximately 2.4 GB compressed. During image push, the Doc
 
 #### Container Runtimes
 
-| Platform | Supported runtimes | Notes |
-|----------|--------------------|-------|
-| Linux | Docker | Primary supported path. |
-| macOS (Apple Silicon) | Colima, Docker Desktop | Install Xcode Command Line Tools (`xcode-select --install`) and start the runtime before running the installer. |
-| macOS (Intel) | Docker Desktop | Start the runtime before running the installer. |
-| Windows WSL | Docker Desktop (WSL backend) | Supported target path. |
-| DGX Spark | Docker | Use the standard installer and `nemoclaw onboard`. |
+| Platform | Supported runtimes | Status | Notes |
+|----------|--------------------|--------|-------|
+| Linux | Docker | Supported | Primary supported path. |
+| macOS (Apple Silicon) | Colima, Docker Desktop | Caveated | Install Xcode Command Line Tools (`xcode-select --install`) and start the runtime before running the installer. |
+| macOS (Intel) | Docker Desktop | Caveated | Start the runtime before running the installer. |
+| Windows WSL2 | Docker Desktop (WSL backend) | Out of scope | WSL2 hosts are not part of the supported NemoClaw target matrix. |
+| Windows native | N/A | Out of scope | Native Windows hosts are not part of the supported NemoClaw target matrix. |
+| Jetson | N/A | Out of scope | Jetson hosts are not part of the supported NemoClaw target matrix. |
+| DGX Spark | Docker | Supported | Use the standard installer and `nemoclaw onboard`. |
 
 ### Install NemoClaw and Onboard OpenClaw Agent
 

diff --git a/docs/about/overview.md b/docs/about/overview.md
@@ -39,14 +39,25 @@ By combining powerful open source models with built-in safety measures, NemoClaw
 
 NemoClaw provides the following capabilities on top of the OpenShell runtime.
 
-| Feature | Description |
-|---------|-------------|
-| Guided onboarding | Validates credentials, selects providers, and creates a working sandbox in one command. |
-| Hardened blueprint | A security-first Dockerfile with capability drops, least-privilege network rules, and declarative policy. |
-| State management | Safe migration of agent state across machines with credential stripping and integrity verification. |
-| Messaging bridges | Host-side processes that connect Telegram, Discord, and Slack to the sandboxed agent. |
-| Routed inference | Provider-routed model calls through the OpenShell gateway, transparent to the agent. Supports NVIDIA Endpoints, OpenAI, Anthropic, Google Gemini, and local Ollama. |
-| Layered protection | Network, filesystem, process, and inference controls that can be hot-reloaded or locked at creation. |
+:::{list-table}
+:header-rows: 1
+
+* - Feature
+  - Description
+* - Guided onboarding
+  - Validates credentials, selects providers, and creates a working sandbox in one command.
+* - Hardened blueprint
+  - A security-first Dockerfile with capability drops, least-privilege network rules, and declarative policy.
+* - State management
+  - Safe migration of agent state across machines with credential stripping and integrity verification.
+* - Messaging bridges
+  - Host-side processes that connect Telegram, Discord, and Slack to the sandboxed agent.
+* - Routed inference
+  - Provider-routed model calls through the OpenShell gateway, transparent to the agent.
+    Supports NVIDIA Endpoints, OpenAI, Anthropic, Google Gemini, compatible OpenAI or Anthropic endpoints, and the caveated Local Ollama path.
+* - Layered protection
+  - Network, filesystem, process, and inference controls that can be hot-reloaded or locked at creation.
+:::
 
 ## Challenge
 

diff --git a/docs/get-started/quickstart.md b/docs/get-started/quickstart.md
@@ -57,13 +57,15 @@ The sandbox image is approximately 2.4 GB compressed. During image push, the Doc
 
 ### Container Runtimes
 
-| Platform | Supported runtimes | Notes |
-|----------|--------------------|-------|
-| Linux | Docker | Primary supported path. |
-| macOS (Apple Silicon) | Colima, Docker Desktop | Install Xcode Command Line Tools (`xcode-select --install`) and start the runtime before running the installer. |
-| macOS (Intel) | Docker Desktop | Start the runtime before running the installer. |
-| Windows WSL | Docker Desktop (WSL backend) | Supported target path. |
-| DGX Spark | Docker | Use the standard installer and `nemoclaw onboard`. |
+| Platform | Supported runtimes | Status | Notes |
+|----------|--------------------|--------|-------|
+| Linux | Docker | Supported | Primary supported path. |
+| macOS (Apple Silicon) | Colima, Docker Desktop | Caveated | Install Xcode Command Line Tools (`xcode-select --install`) and start the runtime before running the installer. |
+| macOS (Intel) | Docker Desktop | Caveated | Start the runtime before running the installer. |
+| Windows WSL2 | Docker Desktop (WSL backend) | Out of scope | WSL2 hosts are not part of the supported NemoClaw target matrix. |
+| Windows native | N/A | Out of scope | Native Windows hosts are not part of the supported NemoClaw target matrix. |
+| Jetson | N/A | Out of scope | Jetson hosts are not part of the supported NemoClaw target matrix. |
+| DGX Spark | Docker | Supported | Use the standard installer and `nemoclaw onboard`. |
 
 ## Install NemoClaw and Onboard OpenClaw Agent
 

diff --git a/docs/inference/inference-options.md b/docs/inference/inference-options.md
@@ -41,15 +41,15 @@ The onboard wizard presents the following provider options by default.
 The first six are always available.
 Ollama appears when it is installed or running on the host.
 
-| Option | Description | Curated models |
-|--------|-------------|----------------|
-| NVIDIA Endpoints | Routes to models hosted on [build.nvidia.com](https://build.nvidia.com). You can also enter any model ID from the catalog. Set `NVIDIA_API_KEY`. | Nemotron 3 Super 120B, Kimi K2.5, GLM-5, MiniMax M2.5, GPT-OSS 120B |
-| OpenAI | Routes to the OpenAI API. Set `OPENAI_API_KEY`. | `gpt-5.4`, `gpt-5.4-mini`, `gpt-5.4-nano`, `gpt-5.4-pro-2026-03-05` |
-| Other OpenAI-compatible endpoint | Routes to any server that implements `/v1/chat/completions`. The wizard prompts for a base URL and model name. Works with OpenRouter, LocalAI, llama.cpp, or any compatible proxy. Set `COMPATIBLE_API_KEY`. | You provide the model name. |
-| Anthropic | Routes to the Anthropic Messages API. Set `ANTHROPIC_API_KEY`. | `claude-sonnet-4-6`, `claude-haiku-4-5`, `claude-opus-4-6` |
-| Other Anthropic-compatible endpoint | Routes to any server that implements the Anthropic Messages API (`/v1/messages`). The wizard prompts for a base URL and model name. Set `COMPATIBLE_ANTHROPIC_API_KEY`. | You provide the model name. |
-| Google Gemini | Routes to Google's OpenAI-compatible endpoint. Set `GEMINI_API_KEY`. | `gemini-3.1-pro-preview`, `gemini-3.1-flash-lite-preview`, `gemini-3-flash-preview`, `gemini-2.5-pro`, `gemini-2.5-flash`, `gemini-2.5-flash-lite` |
-| Local Ollama | Routes to a local Ollama instance on `localhost:11434`. NemoClaw detects installed models, offers starter models if none are present, pulls and warms the selected model, and validates it. | Selected during onboarding. For more information, refer to [Use a Local Inference Server](use-local-inference.md). |
+| Option | Status | Description | Curated models |
+|--------|--------|-------------|----------------|
+| NVIDIA Endpoints | Supported | Routes to models hosted on [build.nvidia.com](https://build.nvidia.com). You can also enter any model ID from the catalog. Set `NVIDIA_API_KEY`. | Nemotron 3 Super 120B, Kimi K2.5, GLM-5, MiniMax M2.5, GPT-OSS 120B |
+| OpenAI | Supported | Routes to the OpenAI API. Set `OPENAI_API_KEY`. | `gpt-5.4`, `gpt-5.4-mini`, `gpt-5.4-nano`, `gpt-5.4-pro-2026-03-05` |
+| Other OpenAI-compatible endpoint | Supported | Routes to any server that implements `/v1/chat/completions`. The wizard prompts for a base URL and model name. Works with OpenRouter, LocalAI, llama.cpp, or any compatible proxy. Set `COMPATIBLE_API_KEY`. | You provide the model name. |
+| Anthropic | Supported | Routes to the Anthropic Messages API. Set `ANTHROPIC_API_KEY`. | `claude-sonnet-4-6`, `claude-haiku-4-5`, `claude-opus-4-6` |
+| Other Anthropic-compatible endpoint | Supported | Routes to any server that implements the Anthropic Messages API (`/v1/messages`). The wizard prompts for a base URL and model name. Set `COMPATIBLE_ANTHROPIC_API_KEY`. | You provide the model name. |
+| Google Gemini | Supported | Routes to Google's OpenAI-compatible endpoint. Set `GEMINI_API_KEY`. | `gemini-3.1-pro-preview`, `gemini-3.1-flash-lite-preview`, `gemini-3-flash-preview`, `gemini-2.5-pro`, `gemini-2.5-flash`, `gemini-2.5-flash-lite` |
+| Local Ollama | Caveated | Routes to a local Ollama instance on `localhost:11434`. NemoClaw detects installed models, offers starter models if none are present, pulls and warms the selected model, and validates it. | Selected during onboarding. For more information, refer to [Use a Local Inference Server](use-local-inference.md). |
 
 ## Experimental Options